J . Am. Ckem. SOC.1984, 106, 3037-3039
C.8
-
I
iI I
--:
*
I IO
4-
2C
30
/
40
50
60
'O4
/
Figure 1. Dependence of equilibrium (left) and kinetics (right) of hydride formation from Co(bpy),+on H' and bpy concentrations. The absorbing species is Co(bpy),' (at (0.3-1.0) X IO" M). 0, 0.02 M Co(II), 0.002 M total bpy (1.2 X IO-' M free bpy), 0.1 M ionic strength (no equilibrium data); A, 0.005 M Co(II), 0.0016 M total bpy (4.6 X lo-' M free bpy), 0.04 M ionic strength; 0 , 0.001 M Co(II), 0.001 M total bpy (2.5 X 10" M free bpy), 0.03 M ionic strength; W, 0.001 M Co(II), 0.002 M total bpy (1.2 X M free bpy), 0.03 M ionic strength. All solutions contained 0.02 M acetateacetic acid and were deoxygenated with argon. The 1:l Co(I1) to bpy solution contained 0.3 M ethanol; the others contained 0.26 M 2-propanol. (Note: k o w values at high [bpy] and [H'] (upper right-hand corner) have not been corrected for the fact that under these conditions eq 1 is comparable in rate to eq 2.)
centration of free bipyridine. Remarkably, however, the slopes also increase with [bpy] and the rate law for the equilibration is given by eq 3 with a = 1.O X 1O7 M-I s-l a nd b = 5.0 X 1OI2M-, kobsd
= ( a + b[bpyl)([H+l
+ ([bpyl/KZ))
fast
+ H + e (bpy)H+
CO(bPY),+ + (bPY)H+ (bpy)H.
(4)
slow
Co(bPY)32++(bPY)H.
+ H+
(bpy)H,+.
(5)
(6)
(bpy)H./ (bpy) H2+-+ Co(bpy)'+/Co( bpy)z2+% C O ( ~ P Y ) Z ( H Z O ) (7) H~+ (bpy)H- (eq 5) is the rate-determining step in the forward direction. In terms of this mechanism, b = K4k5and the b values used in calculating the lines in Figure 1 were obtained from the values reported previ~usly:~ K4 = 2.6 X lo4 M-I, kS = 1.8 X lo8 M-' s-' (K6 = 1 X lo8 M-'). Although the a term in the rate law is of the form expected if eq 2 is an elementary reaction (i.e., k , = 1.0 X lo7 M-I s-l 1, the fact that (bpy)H. production is implicated for the b term suggests an analogous process for the a term, Le., eq 4 and 6-9, in which rapid preequilibrium eq 8 is maintained by the reverse fast
CO(bPY),+
CO(bPY)2+ + bPY
electron exchange8s9(- 1 X lo9 M-' s-' ). Thus K4K8k9is estimated as 1 X lo7 M-I s-l , a nd this route is sufficiently rapid to account quantitatively for the a term. Thus conventional B r ~ n s t e dproton-transfer paths are not detected in this system. They could, nevertheless, be relatively rapid: our observations impose a limit of < l o 7 M-' s-I for reaction of either Co(bpy),+ or Co(bpy),+ with H 3 0 + . The above considerations suggest that the generation of (bpy)H. via reduction of (bpy)H+ by Co(bpy),+ or Co(bpy),+ is the rate-determining step in Co(bpy),(H20)H2' formation. Thus actual assemblage of the hydride-presumably via reaction of Co(bpy),+ or Co(bpy)? with (bpy)H. or (bpy)H2+. (eq 7)-must be extremely facile. The reaction could involve H atom transfer or sequential electron and proton transfer, with either possibly coupled to substitution of (bpy)H. on Co(I1). Finally, the generality of the free radical route to hydrides is of some interest. Such routes likely obtain in other Co(1) polypyridine systems, being facilitated by the similarity of the reduction potentials for C O L T / + and LH+lo couple^,^ the rapidity of electron transfer among these couples, and (probably) the relatively high substitutional lability of the Co(I1) species.I3 Whether or not such routes prevail with other metal centers or other reducible ligands remains to be demonstrated. Acknowledgment. This work was performed at Brookhaven National Laboratory under the auspices of the U.S. Department of Energy and supported by its Office of Basic Energy Sciences. (1 3) Hydride formation mechanisms in other Co(1) polypyridine systems are currently under study as are the routes via which the hydrides react with water to give H2.Hydride formation via H-atom abstraction from organic radicals finds precedent in the C~(CN),~-/alkyl halide systems (Chock, P. B.; Halpern, J. H. J . Am. Chem. Soc. 1969, 91, 582).
(3)
s-l. The reaction proceeds by parallel ( a and 6 ) paths. The two terms in [H+] are the forward components of the rate of approach to equilibrium while the two terms in [bpy]/K, are from the reverse rates. The magnitude and concentration dependence of the b term in eq 3 suggests the sequence eq 4-712 in which the formation of
bpy
3037
(8)
=
A d(GpG)-Platinated Oligonucleotide Can Form a Duplex with a Complementary Strand Bruno Van Hemelryck,Ia Eric Guittet,Ib Genevitve Chottard,Ic Jean-Pierre Girault,'" Tam Huynh-Dinh,ld Jean-Yves Lallemand,lb Jean Igolen,Id and Jean-Claude Chottard*Ia Laboratoire de Chimie de I'Ecole Normale SupPrieure 75231 Paris Cedex 05, France Institut de Chimie des Substances Naturelles 91 190 Gif-sur- Yvette, France UniversitP Pierre et Marie Curie 75230 Paris Cedex 05, France Institut Pasteur, 75724 Paris Cedex 15, France Receioed December 19. 1983
In the cell, DNA appears to be the primary target of the active aquated forms of the antitumor drug cis-[PtCI,(NH,),] (cisDDP),' and the cytotoxicity of the drug could result of a particular bifunctional binding of the cis-Pt"(NH,), moiety., It has been shown, using enzymatic digestion methods, that platinum crosslinks between adjacent guanines are formed upon reaction of DNA with ~ i s - D D pand , ~ represent more than 50% of the lesion^.^ This is in agreement with the results obtained by enzymatic restriction
slow
CO(~PY),+ + (~PY)H+ Co(bpy)z2++ (bpy)H* (9) of eq 1. In this mechanism a = K4K8k9. The magnitude of K8 = 1.3 X lo-' M, and k9 is estimated as 3 X IO9 M-' s-l from the Co(bpy)?+/+ and (bpy)H+ioEo values (-1.038 and -0.97 V,9 respectively) and the fact that both couples undergo very rapid (12) Given the buffer and Co(I1) concentrations used, eq 6 is probably rapid compared to eq 7 . Thus reaction of either (bpy)H. or (bpy)H2+.with Co(I1) (eq 7, Co(bpy)2+ or Co(bpy)2Z+)is a possibility.
0002-7863/84/ 1506-3037$01.50/0
(1) (a) Laboratoire de chimie de wrdination organique et biologique, LA 32. (b) Laboratoire de rCsonance magnBtique nucltaire. (c) DBpartement des recherches physiques, LA 71. (d) Unite de chimie organique ERA 927. (2) (a) Roberts, J. J.; Thomson, A. J. Prog. Nucleic Acid Res. Mol. B i d . 1979, 22, 71 and references therein. (b) Roberts, J. J.; Pera M.F., Jr. ACS Symp. Ser. 1983, 209, 3 and references therein. (3) Kelman, A. D.; Peresie, H. J. Cancer Treat. Rep. 1979, 63, 1445 and references therein. (4) Fichtinger-Schepman, A.-M.; Lohman, P. H. M.; Reedijk, J. Nucl. Acids Res. 1982, 10, 5345. (5) Eastman, A. Biochemistry 1983, 22, 3927.
0 1984 American Chemical Society
3038 J . Am. Chem. SOC.,Vol. 106, No. 10, 1984 1CPtI
-
Communications to the Editor
+2
24%
1 + 2 44%
/
A
~
I
/'
/
/f
'X
Figure 1. (a) Variation of the normalized absorbance (absorbances are
+
normalized by dividing by the absorbance at 80 "C) of (1 2 ) at 258 nm (-*-) and ( l [ P t ] 2 ) at 260 nm (-) vs. temperature (actual temperature progression: 0.5 OC per min). Solutions: cacodylate buffer 8 m M p H 7.6, NaCl 1 M , EDTA 2 m M ; (1 2 ) 2.3 X M , (1[Pt] 2) 8 X 10" M oligomer strand concentrations. (b) Variation of the relative value of Ac((Aco - Acr)/Aco) (Ace and AtT are the differential absorptions at 0 and T OC respectively) at 285 nm vs. temperature of ( 1 2 and ( l [ P t ] 2) (-). Solutions: cacodylate buffer 10 m M p H 7, NaCl 0.2 M ; (1 2) 1.8 X M, (l[Pt] 2) 2 X M oligomer strand concentrations.
+
+
+
+
(-e-)
+
+
+
1%
5%
K ppm 14
13
i2
7
14
13
12
techniques6 and by the studies of oligonucleotide models, which pointed to the facile N7-N7 platinum chelation by two adjacent 3' "I 2 3 4 5 6 7 8 guanines.'-lo Many studies of the perturbation of DNA secondary d/GATCCGGC) structure and stability, induced by the binding of the C ~ S - P ~ " ( N H ~ ) ~ moiety, have revealed a strong distortion of the DNA duplex suggesting the disruption of several base pairs."-14 I t is known 3( 18 CG 17 C 1615141312 TAGGC 11 C 10 G 9 j5,d that the d(GpG)-chelated C ~ S - P ~ I I ( N adducts H ~ ) ~ of the selfcomplementary hexanucleotides d(A-G-G-C-C-T) and d(T-GFigure 2. 12-14.5 ppm region of the 400-MHz ' H N M R spectra of (1 G-C-C-A) do not form a duplex s t r u ~ t u r e . ~We ~ ' ~here report + 2 ) (right) and (1[Pt] + 2 ) (left) at different temperatures. Chemical preliminary results which show that the d(GpG)-platinum adduct shifts are relative to TSPd4 (sodium 3-(trimethylsilyl)-l-propionateof the octanucleotide d(G-A-T-C-C-G-G-C) ( l ) , cis-[Pt2,2,3,3-d4). Samples: phosphate buffer 10 m M p H 6.9 in H,0/D20 (NH3),(d(G-A-T-C-C-G-G-C)-N7(6),N7(7))] 5- hereafter called 90/10, NaC10.2 M; (1 + 2) 2.3 X lo-) M , ( l [ P t ] + 2) 8 X lo4 M. The spectra were recorded on a Bruker WM-400, by using a time-shared 1[Pt], gives a duplex with the complementary decanucleotide "soft" pulse to reduce the water signal.25 The arrows indicate the 13.7, d(G-C-C-G-G-A-T-C-G-C) (2). The melting temperature of the 13.6, and 13.30 ppm signals first affected by the temperature increase. M) is 30 "C and there are eight imino (1[Pt] 2) duplex The numbering of the bases is indicated at the bottom of the figure. protons involved in hydrogen bonding between the two strands. The deoxyoligonucleotides 1 and 2 were synthesized from dimers ~ ' ~ *(i)' ~ by a phosphotriester method in solution, as reported p r e v i o u ~ l y . ~ ~ is unambiguously assigned by the usual r n e t h o d ~ ; ~i.e.: high-pressure gel permeation chromatographyi0and atomic abThe 1[Pt] adduct was prepared by the stoichiometric reaction of sorption spectroscopy coupled with the UV absorption of the with the octamer 1 in cis-DDP or C~S-[P~(NH~)~(H~O),](NO~)~ complex1*show that 1[Pt] is monomeric and contains one platinum previously described conditions.1° 1[Pt] was the major complex atom; (ii) 400-MHz ' H N M R allows the assignment of the formed and was separated by preparative HPLC.16 Its structure platinum binding sites from the analysis of the pH dependence of the chemical shifts of the nonexchangeable base protons," in agreement with the H8 downfield shifts of the coordinated gua(6) (a) Cohen, G. L.; Ledner, J. A.; Bauer, W. R.; Ushay, H. M.; Cara-
1
2
+
vans, C.; Lippard, S. J. J . Am. Chem. SOC.1980,102, 2487. (b) Ushay, H. M.; Tullius, T. D.; Lippard, S.J. Biochemistry 1981, 20, 3744. (c) Tullius, T. D.; Lippard, S.J. J. Am. Chem. SOC.1981, 103, 4620. (d) Tullius, T. D.; Ushay, H. M.; Merkel, C. N.; Caradonna, J. P.; Lippard, S.J. ACS Symp. Ser. 1983, 209, 5 1. (7) (a) Chottard, J.-C.; Girault, J.-P.; Chottard, G.; Lallemand, J.-Y.; Mansuy, D. Nouu. J . Chim. 1978, 2, 551. (b) Chottard, J.-C.; Girault, J.-P.; Chottard, G.; Lallemand, J.-Y.; Mansuy, D. J . Am. Chem. SOC.1980, 102, 5565. (c) Girault, J. P.; Chottard, G.; Lallemand, J.-Y.; Chottard, J.-C. Biochemistry 1982, 21, 1352. (d) Chottard, J.-C.; Girault, J.-P.; Guittet, E. R.; Lallemand, J.-Y.; Chottard, G. ACS Symp. Ser. 1983, 209, 125. (8) (a) Marcelis, A. T. M.; Canters, G.W.; Reedijk, J. Recl. Trua. Chim. Puys-Bus 1981,100, 391. (b) Marcelis, A. T. M.; Reedijk, J. 1983, 102, 121. (9) Caradonna, J.-P.; Lippard, S. J. J . Am. Chem. SOC.1982, 104, 5793. (10) Girault, J.-P.; Chottard, J.-C; Guittet, E. R.; Lallemand, J.-Y.; Huynh-Dinh, T.: Igolen, J. Biochem. Biophys. Res. Commun. 1982, 109, 1157. (11) (a) Macquet, J.-P.; Butour, J.-L. Biochimie 1978, 60, 901. (b) Macquet, J.-P.; Butour, J.-L.; Johnson, N. P. ACSSymp. Ser. 1983, 209, 75. (12) Srivasta, R. C.; Froehlich, J.; Eichhorn, G. L. Biochimie 1978, 60, 879. (13) (a) Cohen, G. L.; Bauer, W. R.; Barton, J. K.; Lippard, S. J. Science (Wushingron, D.C.)1979, 203, 1014. (b) Lippard, S. J. ACS Symp. Ser. 1980, 140, 147. (e) Reference 6b. (14) Scovell, W. M.; Capponi, V . J. Biochem. Biophys. Res. Commun. 1982, 107, 11 38. (15) Tran-Dinh, S.;Neumann, J. M.; Huynh-Dinh, T.; Lallemand, J.-Y.; Igolen, J. Nucl. Acids Res. 1982, 10, 5319.
nines.8-'0J7
In order to assess the possibility of duplex formation between the platinated octamer 1[Pt] and the complementary decamer 2, (16) Reverse-phase HPLC analyses were performed on a Beckman 421 liquid chromatogra h with 254-nm detection, on a Waters C18 fi-Bondapak column, using a 10- M aqueous CH,CO,NH, solution as eluant A and a IO-* M CH3C02NH4solution in H20/CH,0H/CH,CN (5:4:1) as eluant B. Both solutions were brought to pH 7.0 by ",OH addition. The preparative separations were performed in the same conditions. (17) Girault, J.-P.; Chottard, J.-C.; Neumann, J.-M.; Tran-Dinh, S.; Huynh-Dinh, T.; Igolen, J. Nouu. J . Chim. 1984, 8, 7. (18) From the UV absorption of the adduct, an approximate c-/Pt molar extinction coefficient of 64000 was calculated. This value compared to that of -60000 for the unplatinated octamer shows that one platinum is bound per oligonucleotide molecule. (19) Among the four purine H8,three are exchanged with D 2 0 at basic pH. The plots of the chemical shifts vs. pH show that the lower field one, experiencing the fastest exchange, belongs to the N7-platinated G(7) (pK,(Nl-H) = 8)" and the others to the N7-platinated G(6) (pK,(Nl-H) = 8.5) and the free G ( l ) (pK,(NI-H) = 9.7). The C-H6, A-H8, and A-H2 curves show the protonation of the free C-N3 and A-N1 sites (pK,(N3) = 4.3, pKa(N1) = 3.5). Platinum binding by the 1 G(1), giving a loop, can be excluded because the adjacent adenosine H8, H2, and HI' are respectively little and not affected by the metal chelation.
P
3039
J . A m . Chem. SOC.1984, 106, 3039-3041
melting occurs at a temperature lower than 37 "C, but the duplex we investigated the effect of temperature on the (1[Pt] + 2) and is quite short, and the platinated sequence is next to the helix end. (1 2) equimolar mixtures. This suggests that d(GpG)-platinum chelation should not induce Both relative absorbance vs. temperature curves for (1 + 2) a large distortion of the DNA duplex. This distortion might be and ( l [ P t ] 2) (Figure l a ) present a sigmoidal shape, with quite similar to that of a kink30 or of a region of systematically respective melting temperatures of 5 5 and 28 "C at 2.3 X 10-5 bent B DNA,,' and this would raise the question of its detection and 8 X lo6 M oligomer strand concentrations. (l[Pt] + 2) shows by the repair enzymes.32 less cooperativity than (1 2). That the observed melting profile of (1 [Pt] + 2) does not reflect the partial self-pairing of any of Acknowledgment. Support of this work by a grant from the the two strands is ascertained by comparison with the curves of "Association pour le d6veloppement de la Recherche sur le Cancer" 1[Pt] and 2 alone.20 (ARC) is gratefully acknowledged. We thank Dr. Toulm6, LaThe circular dichroism spectra of (1 + 2) and (l[Pt] + 2) boratoire de Biophysique, Mus6um d'Histoire Naturelle, Paris, belong to the B DNA the latter showing a larger differential and Dr. Gouyette, Institut Gustave Roussy, Villejuif, for the access absorption in the 270-290-nm region. The relative At2,, vs. to their UV and AAS facilities. We thank Engelhardt Industries, temperature plots for (1 + 2) and (l[Pt] 2) (Figure lb) show France, for a loan of platinum salt. respective melting temperatures of 50 and 30 "C at 1.8 X lo-' and 2 X 10-5 M oligomer strand concentrations. These T , values are in agreement with those of the UV melting profiles. Here (30) Sobell, H. M.; Tsai, C. C.; Jain, S. C.; Gilbert, S. G. J . Mol. Bid. 1977, 114, 333. too, none of the Atzss vs. temperature curves for 1[Pt] and 2 alone (31) Marini, J.-C.; Levene, S. D.; Crothers, D. M.; Englund, P. T. Proc. exhibit a sigmoidal profile.22 Natl. Acad. Sci. U.S.A. 1982, 79. 7664. The N M R spectra of the exchangeable imino protons (G-N1H (32) We thank professor J.-B. Le Pecq for a discussion of this point. and T-N3H) of (1 2) and (1[Pt] + 2) in water23,24are shown in Figure 2 at different temperatures. For (1 2) the integration gives eight imino protons. The lower field group (2 H ) is assigned to the two A-T base pairs (2-15 and 3-14, see Figure 2 for the numbering) based on the nuclear Overhauser effects between these protons and the two A-H2. The two higher field groups of signals Measurement of the Thermochromic Equilibrium (4H and 2H) belong to the G-C base pairs.26 All these signals Constant of a Nonthermochromic Compound: are little affected by raising the temperature up to 40 "C. But 1,l'-Dimethylbianthrone at 44 "C, they experience a general broadening and disappear at 50 OC. For (l[Pt] 2) at low temperature, the signal pattern is completely different. It is noteworthy that there are still eight Dennis H. Evans* and Alanah Fitch imino protons involved in hydrogen bonding.26 Compared to (1 Department of Chemisrry 2), four G-C imino protons are moved: three are downfield University of Wisconsin-Madison shifted, two overlapping with the A-T's, and one is upfield shifted, Madison, Wisconsin 53706 overlapping with the two high field G-C's. Raising the temperature from 1 to 24 "C affects the three signals at 13.7, 13.6, and Received January 4, 1984 13.30 ppm. At 30 "C all the signals have disappeared, and this Revised Manuscript Received March 24, 1984 agrees with the lower UV and CD T , obtained for (1[Pt] 2) compared to (1 2). Three G-C imino protons of (l[Pt] 2) Bianthrone [Chemical Abstracts name: I O - ( 1 O-oxo-9are involved in fraying of the duplex, which should first affect (lOH)-anthracenylidene)-9( lOtl)-anthracenone ( l ) ] is a therthe (6-1 I ) , (7-IO), and (8-9) base pairs of the platinated end if one takes into account the stabilization of the other end by the 0 unpaired G ( 17) and C( 18) bases of the decamer ~ t r a n d . ~ ' * ~ ~ Therefore the three downfield shifted G-C imino protons should correspond to the G-platinated (6-1 1) and (7-10) pairs and to the distorted terminal (8-9) pair. The upfield shifted G-C imino proton could be assigned to the (5-12) pair because of the perturbation due to the adjacent platinum chelate. These assignments are tentative and are in agreement with preliminary results obtained by other workers.29 In conclusion, our preliminary results show that a d(GpG)0 c i ~ - p t ( N H , )chelate ~ on a short oligonucleotide does not preclude A B duplex formation with a complementary strand. In our case,
+
+
+
+
+
+
+
+
+
+ +
1 (20) That of l[Pt] shows a small increase of the absorbance with temperature followed by a decrease above 28 OC and that of 2 shows a nearly linear increase with temperature. (21) Woody, R. W. Macromol. Reo. 1977, 12, 181. (22) That of l[Pt] shows an immediate sharp increase with temperature up to 30 OC followed by a smaller one and no variation above 55 OC; that of 2 shows a nearly linear increase with temperature. (23) Hilbers, C. W. 'Biological Applications of Magnetic Resonance"; Schulman, R. G., Ed.; Academic Press: New York, 1979; pp 1-43. (24) Patel, D. J.; Pardi, A,; Itakura, K. Sience (Washington, D.C.)1982, 216, 581. (25) Morris, G . A.; Freeman, R. J . Magn. Reson. 1978, 29, 433. (26) At the temperatures used we have no evidence for end-to-end agregation of the (1 + 2) or (l[Pt] + 2) duplexes through hydrogen bond formation involving G(17) and C(18).27 (27) Patel, D. J.; Kozlowski, S. A.; Marky, L. A.; Rice, J. A,; Broka, C.; Itakura, K.; Breslauer, K. J. Biochemistry 1982, 21, 451. (28) Petersheim, M.; Turner, D. H . Biochemistry 1983, 22, 256. (29) In the case of the duplex formed between the d(GpG)-platinated d(T-C-T-C-G-G-T-C-T-C) strand and the complementary d(G-A-G-A-C-CG-A-G-A). Den Hartog, J . H. J.; Altona, C.; van Boom, J. H.; van der Marel, G. A.; Haasnoot, C. A. G.; Reedijk, J. J . A m . Chem. Sot. 1984, 106, 1528.
0002-7863/84/ 1506-3039$01.50/0
mochromic substance that exists at room temperature as a yellow A form. As solutions of 1 are heated, a significant fraction is converted to the green B form whose enthalpy is 3.0 kcal/mol greater than A,' Substituents larger than fluorine at the 1- and 1'-positions prevent thermochromicity.2 It is likely that bulky substituents destabilize B more than A causing the energy difference to be too high to allow formation of detectable quantities of B at temperatures below decompo~ition.~Until now no means of measuring this energy difference has been available. We have developed an indirect electrochemical technique and have found KAB= [B]/[A] = 8 X at 373 K (AG,,, = 8.7 kcal/mol) ( 1 ) Tapuhi, Y.; Kalisky, 0.; Agranat, I . J . Org. Chem. 1979, 4 4 , 1949-1 952. (2) Day, J. H . Chem. Reo. 1963, 63, 65-80. (3) (a) Harnik, E. J . Chem. Phys. 1956, 24, 297-299. (b) Korenstein, R.; Muszkat, K. A,; Sharafy-Ozeri, S. J . A m . Chem. SOC.1973, 95, 6177-6181. .. . .. .-
0 1984 American Chemical Society
